rsshub 1.0.0-master.f992838 → 1.0.0-master.f9a482d

package/lib/v2/oeeee/app/channel.js

@@ -0,0 +1,36 @@
+const got = require('@/utils/got');
+const timezone = require('@/utils/timezone');
+const { parseDate } = require('@/utils/parse-date');
+const { parseArticle } = require('../utils');
+const { art } = require('@/utils/render');
+const path = require('path');
+
+module.exports = async (ctx) => {
+    const id = ctx.params.id ?? 50;
+    const currentUrl = `https://api-ndapp.oeeee.com/friends.php?m=Zone&a=SpaceDoclist&uid=${id}&type=doc`;
+
+    const { data: response } = await got(currentUrl);
+
+    const list = response.data
+        .filter((i) => i.url) // Remove banner and sticky articles.
+        .map((item) => ({
+            title: item.title,
+            description: art(path.join(__dirname, '../templates/description.art'), {
+                thumb: item.titleimg.replace(/\?x-oss-process=.*/g, ''),
+                description: item.summary,
+            }),
+            pubDate: timezone(parseDate(item.ptime * 1000), +8),
+            link: item.url,
+            channel: item.author,
+        }));
+
+    const channel = list[1] ? list[1].channel : '';
+
+    const items = await Promise.all(list.map((item) => parseArticle(item, ctx.cache.tryGet)));
+
+    ctx.state.data = {
+        title: `南方都市报客户端 - ${channel}`,
+        link: `https://m.mp.oeeee.com/u/${id}.html`,
+        item: items,
+    };
+};

package/lib/v2/oeeee/app/reporter.js

@@ -0,0 +1,30 @@
+const got = require('@/utils/got');
+const { parseArticle } = require('../utils');
+const { art } = require('@/utils/render');
+const path = require('path');
+
+module.exports = async (ctx) => {
+    const id = ctx.params.id ?? 0;
+    const currentUrl = `https://m.mp.oeeee.com/show.php?m=Doc&a=getAuthorInfo&id=${id}`;
+
+    const { data: response } = await got(currentUrl);
+
+    const list = response.data.list.map((item) => ({
+        title: '【' + item.media_nickname + '】' + item.title,
+        description: art(path.join(__dirname, '../templates/description.art'), {
+            thumb: item.titleimg,
+            description: item.summary,
+        }),
+        link: item.url,
+    }));
+
+    const author = response.data.info ? response.data.info.name : '';
+
+    const items = await Promise.all(list.map((item) => parseArticle(item, ctx.cache.tryGet)));
+
+    ctx.state.data = {
+        title: `南方都市报奥一网 - ${author}`,
+        link: `https://m.mp.oeeee.com/w/${id}.html`,
+        item: items,
+    };
+};

package/lib/v2/oeeee/maintainer.js

@@ -0,0 +1,5 @@
+module.exports = {
+    '/web/:channel': ['TimWu007'],
+    '/app/channel/:id': ['TimWu007'],
+    '/app/reporter/:id': ['TimWu007'],
+};

package/lib/v2/oeeee/radar.js

@@ -0,0 +1,12 @@
+module.exports = {
+    'oeeee.com': {
+        _name: '奥一网',
+        '.': [
+            {
+                title: '新闻',
+                docs: 'https://docs.rsshub.app/traditional-media.html#nan-fang-du-shi-bao',
+                source: ['/'],
+            },
+        ],
+    },
+};

package/lib/v2/oeeee/router.js

@@ -0,0 +1,5 @@
+module.exports = (router) => {
+    router.get('/web/:channel', require('./web'));
+    router.get('/app/channel/:id', require('./app/channel'));
+    router.get('/app/reporter/:id', require('./app/reporter'));
+};

package/lib/v2/oeeee/templates/description.art

@@ -0,0 +1,7 @@
+{{ if thumb }}
+<img src="{{ thumb }}"><br>
+{{ /if }}
+{{ if description }}
+    <blockquote><p>{{ description }}</p></blockquote>
+    <br>
+{{ /if }}

package/lib/v2/oeeee/utils.js

@@ -0,0 +1,41 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const timezone = require('@/utils/timezone');
+const { parseDate } = require('@/utils/parse-date');
+
+const parseArticle = (item, tryGet) =>
+    tryGet(item.link, async () => {
+        const detailResponse = await got({
+            method: 'get',
+            url: item.link,
+        });
+
+        const content = cheerio.load(detailResponse.data);
+
+        // nfapp page:
+        if (content('.post-cont') && content('.post-cont').html()) {
+            content('img').removeAttr('data-width').removeAttr('data-height').removeAttr('class').removeAttr('title').removeAttr('referrerpolicy');
+            content('.taglist, .J_ndlogo, .zan-shang, .sourcelist-box, #shareContain, .buyCopyright, .article-info, .icon, .special').remove();
+            item.description +=
+                content('.post-cont')
+                    .html()
+                    .replace(/data:image\S*=="\s*\n*\s*original="/g, '') ?? '';
+            if (!item.pubDate) {
+                item.pubDate = timezone(parseDate(content('.introduce').text().split()), +8);
+            }
+        }
+        // oeeee news page:
+        else if (content('.content') && content('.content').html()) {
+            item.description += content('.content').html() ?? '';
+        }
+        // oeeee news page #2:
+        else if (content('.article-conten') && content('.article-conten').html()) {
+            item.description += content('.article-conten').html() ?? '';
+        }
+
+        return item;
+    });
+
+module.exports = {
+    parseArticle,
+};

package/lib/v2/oeeee/web.js

@@ -0,0 +1,36 @@
+const got = require('@/utils/got');
+const timezone = require('@/utils/timezone');
+const { parseDate } = require('@/utils/parse-date');
+const { parseArticle } = require('./utils');
+const { art } = require('@/utils/render');
+const path = require('path');
+
+module.exports = async (ctx) => {
+    const channel = ctx.params.channel ?? 0;
+    const currentUrl = `https://www.oeeee.com/api/channel.php?m=Js4channelNews&a=newLatest&cid=${channel}`;
+
+    const { data: response } = await got(currentUrl);
+
+    const list = response.data.map((item) => ({
+        title: '【' + item.channel_name + '】' + item.title,
+        description: art(path.join(__dirname, 'templates/description.art'), {
+            thumb: item.img,
+            description: item.summary,
+        }),
+        pubDate: timezone(parseDate(item.datetime), +8),
+        link: item.linkurl,
+        author: item.author,
+        // channelName: item.channel_name,
+        channelEname: item.channel_ename,
+    }));
+
+    const channelEname = list[1] ? list[1].channelEname : '';
+
+    const items = await Promise.all(list.map((item) => parseArticle(item, ctx.cache.tryGet)));
+
+    ctx.state.data = {
+        title: `南方都市报奥一网`,
+        link: `https://www.oeeee.com/api/channel.php?s=/index/index/channel/${channelEname}`,
+        item: items,
+    };
+};

package/lib/v2/oilchem/index.js

@@ -33,29 +33,28 @@ module.exports = async (ctx) => {
         .get();
 
     const items = await Promise.all(
-        list.map(
-            async (item) =>
-                await ctx.cache.tryGet(item.link, async () => {
-                    const detailResponse = await got({
-                        method: 'get',
-                        url: item.link,
-                    });
-                    const content = cheerio.load(detailResponse.data);
+        list.map((item) =>
+            ctx.cache.tryGet(item.link, async () => {
+                const detailResponse = await got({
+                    method: 'get',
+                    url: item.link,
+                });
+                const content = cheerio.load(detailResponse.data);
 
-                    item.description = content('#content').html();
-                    item.pubDate = timezone(
-                        parseDate(
-                            content('.xq-head')
-                                .find('span')
-                                .text()
-                                .match(/发布时间：\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}/)[0],
-                            'YYYY-MM-DD HH:mm'
-                        ),
-                        +8
-                    );
+                item.description = content('#content').html();
+                item.pubDate = timezone(
+                    parseDate(
+                        content('.xq-head')
+                            .find('span')
+                            .text()
+                            .match(/发布时间：\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}/)[0],
+                        'YYYY-MM-DD HH:mm'
+                    ),
+                    +8
+                );
 
-                    return item;
-                })
+                return item;
+            })
         )
     );
 

package/lib/v2/oncc/maintainer.js

@@ -1,3 +1,4 @@
 module.exports = {
     '/:language/:channel?': ['Fatpandac'],
+    '/money18/:id?': ['nczitzk'],
 };

package/lib/v2/oncc/money18.js

@@ -0,0 +1,115 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const dayjs = require('dayjs');
+const timezone = require('@/utils/timezone');
+const { parseDate } = require('@/utils/parse-date');
+const { art } = require('@/utils/render');
+const path = require('path');
+
+const sections = {
+    exp: '新聞總覽',
+    fov: '全日焦點',
+    industry: '板塊新聞',
+    int: '國際金融',
+    recagent: '大行報告',
+    ntlgroup: 'A股新聞',
+    pro: '地產新聞',
+    weainvest: '投資理財',
+    ipo: '新股IPO',
+    tech: '科技財情',
+};
+
+module.exports = async (ctx) => {
+    const id = ctx.params.id ?? 'exp';
+    const limit = ctx.query.limit ? parseInt(ctx.query.limit) : 30;
+
+    const rootUrl = 'https://money18.on.cc';
+    const currentUrl = `${rootUrl}/finnews/news_${id === 'industry' ? 'industry.html' : `breaking.html?section=${id}`}`;
+    const ipoApiUrl = `https://dyn.on.cc/api/asrh/v1/events/names/新股/articles?page=1&limit=${limit}`;
+    const industryApiUrl = `${rootUrl}/cnt/utf8/content/articleList/sector_list_exp_1.js`;
+
+    const toApiUrl = (date) => `${rootUrl}/cnt/utf8/content/${date}/articleList/list_${id}_all.js`;
+
+    let apiUrl = id === 'ipo' ? ipoApiUrl : id === 'industry' ? industryApiUrl : toApiUrl(dayjs().format('YYYYMMDD')),
+        hasArticle = false,
+        items = [],
+        i = 0,
+        response;
+
+    /* eslint-disable no-await-in-loop */
+
+    while (!hasArticle) {
+        try {
+            response = await got({
+                method: 'get',
+                url: apiUrl,
+            });
+            hasArticle = true;
+        } catch (e) {
+            if (e.code === 'ERR_NON_2XX_3XX_RESPONSE') {
+                hasArticle = false;
+                apiUrl = toApiUrl(dayjs().subtract(++i, 'day').format('YYYYMMDD'));
+            }
+        }
+    }
+
+    /* eslint-enable no-await-in-loop */
+
+    if (id === 'ipo') {
+        items = response.data.articles.map((item) => ({
+            title: item.title,
+            author: item.authorname,
+            link: `${rootUrl}/finnews/content/${id}/${item.articleId}.html`,
+            description: art(path.join(__dirname, 'templates/money18.art'), {
+                images: item.hasHdPhoto ? [`https://hk.on.cc/hk/bkn${item.hdEnlargeThumbnail}`] : undefined,
+                description: item.content,
+            }),
+            pubDate: timezone(parseDate(item.pubDate), +8),
+        }));
+    } else if (id === 'industry') {
+        items = response.data.articles.slice(0, limit).map((item) => ({
+            title: item.title,
+            author: item.authorname,
+            link: `${rootUrl}/finnews/content/${id}/${item.articleId}.html`,
+            category: item.sector.map((s) => s.name),
+            pubDate: timezone(parseDate(item.pubDate), +8),
+        }));
+    } else {
+        items = response.data.slice(0, limit).map((item) => ({
+            title: item.title,
+            author: item.authorname,
+            link: `${rootUrl}/finnews/content/${id}/${item.articleId}.html`,
+            pubDate: timezone(parseDate(item.pubDate), +8),
+        }));
+    }
+
+    if (id !== 'ipo') {
+        items = await Promise.all(
+            items.map((item) =>
+                ctx.cache.tryGet(item.link, async () => {
+                    const detailResponse = await got({
+                        method: 'get',
+                        url: item.link,
+                    });
+
+                    const content = cheerio.load(detailResponse.data);
+
+                    item.description = art(path.join(__dirname, 'templates/money18.art'), {
+                        images: content('.photo img')
+                            .toArray()
+                            .map((i) => content(i).attr('src')),
+                        description: content('.content').html(),
+                    });
+
+                    return item;
+                })
+            )
+        );
+    }
+
+    ctx.state.data = {
+        title: `東網產經 - ${sections[id]}`,
+        link: currentUrl,
+        item: items,
+    };
+};

package/lib/v2/oncc/radar.js

@@ -4,34 +4,42 @@ module.exports = {
         hk: [
             {
                 title: '港澳',
-                docs: 'https://docs.rsshub.app/other.html#guo-jia-zi-ran-ke-xue-ji-jin-wei-yuan-hui',
+                docs: 'https://docs.rsshub.app/traditional-media.html#dong-wang',
                 source: ['/hk/news/index.html', '/hk/news/index_cn.html'],
                 target: '/oncc/zh-hans/news',
             },
             {
                 title: '两岸',
-                docs: 'https://docs.rsshub.app/other.html#guo-jia-zi-ran-ke-xue-ji-jin-wei-yuan-hui',
+                docs: 'https://docs.rsshub.app/traditional-media.html#dong-wang',
                 source: ['/hk/cnnews/index.html', '/hk/cnnews/index_cn.html'],
                 target: '/oncc/zh-hans/cnnews',
             },
             {
                 title: '国际',
-                docs: 'https://docs.rsshub.app/other.html#guo-jia-zi-ran-ke-xue-ji-jin-wei-yuan-hui',
+                docs: 'https://docs.rsshub.app/traditional-media.html#dong-wang',
                 source: ['/hk/intnews/index.html', '/hk/intnews/index_cn.html'],
                 target: '/oncc/zh-hans/intnews',
             },
             {
                 title: '评论',
-                docs: 'https://docs.rsshub.app/other.html#guo-jia-zi-ran-ke-xue-ji-jin-wei-yuan-hui',
+                docs: 'https://docs.rsshub.app/traditional-media.html#dong-wang',
                 source: ['/hk/commentary/index.html', '/hk/commentary/index_cn.html'],
                 target: '/oncc/zh-hans/commentary',
             },
             {
                 title: '产经',
-                docs: 'https://docs.rsshub.app/other.html#guo-jia-zi-ran-ke-xue-ji-jin-wei-yuan-hui',
+                docs: 'https://docs.rsshub.app/traditional-media.html#dong-wang',
                 source: ['/hk/finance/index.html', '/hk/finance/index_cn.html'],
                 target: '/oncc/zh-hans/finance',
             },
         ],
+        money18: [
+            {
+                title: '產經',
+                docs: 'https://docs.rsshub.app/traditional-media.html#dong-wang',
+                source: ['/finnews/news_breaking.html'],
+                target: (params, url) => `/oncc/money18/${new URL(url).searchParams.get('section')}`,
+            },
+        ],
     },
 };

package/lib/v2/oncc/router.js

@@ -1,3 +1,4 @@
 module.exports = function (router) {
+    router.get('/money18/:id?', require('./money18'));
     router.get('/:language/:channel?', require('./index'));
 };

package/lib/v2/oncc/templates/money18.art

@@ -0,0 +1,6 @@
+{{ if images }}
+{{ each images image }}
+<img src="{{ image }}">
+{{ /each }}
+{{ /if }}
+{{@ description }}

package/lib/v2/openai/blog.js

@@ -0,0 +1,44 @@
+const got = require('@/utils/got');
+const { toTitleCase } = require('@/utils/common-utils');
+const { getApiUrl, parseArticle } = require('./common');
+
+module.exports = async (ctx) => {
+    const tag = ctx.params.tag || '';
+
+    const rootUrl = 'https://openai.com';
+    const blogRootUrl = 'https://openai.com/blog';
+    const blogOriginUrl = `${rootUrl}/blog${tag === '' ? '' : `?topics=${tag}`}`;
+
+    const apiUrl = new URL('/api/v1/blog-details', await getApiUrl());
+
+    // Construct API query
+    apiUrl.searchParams.append('sort', '-publicationDate,-createdAt');
+    apiUrl.searchParams.append('page[size]', '20');
+    apiUrl.searchParams.append('page[number]', '1');
+    apiUrl.searchParams.append('include', 'media,topics,authors');
+    if (tag) {
+        apiUrl.searchParams.append('filter[topics][slugs][0]', tag);
+    }
+
+    const response = await got({
+        method: 'get',
+        url: apiUrl,
+    });
+
+    const list = response.data.data.filter((entry) => entry.type === 'blog-details');
+
+    const items = await Promise.all(
+        list.map((item) => {
+            const attributes = item.attributes;
+            return parseArticle(ctx, blogRootUrl, attributes);
+        })
+    );
+
+    const title = `OpenAI Blog${tag ? ` - ${toTitleCase(tag)}` : ''}`;
+
+    ctx.state.data = {
+        title,
+        link: blogOriginUrl,
+        item: items,
+    };
+};

package/lib/v2/openai/chatgpt.js

@@ -0,0 +1,98 @@
+const dayjs = require('dayjs');
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const config = require('@/config').value;
+
+dayjs.extend(require('dayjs/plugin/isSameOrBefore'));
+
+module.exports = async (ctx) => {
+    const articleUrl = 'https://help.openai.com/en/articles/6825453-chatgpt-release-notes';
+
+    const cache = await ctx.cache.tryGet(
+        articleUrl,
+        async () => {
+            const returns = [];
+
+            const pageResponse = await got({
+                method: 'get',
+                url: articleUrl,
+            });
+
+            const $ = cheerio.load(pageResponse.data);
+            const page = JSON.parse($('script#__NEXT_DATA__').text()); // 页面貌似是用 Next 渲染的，有现成的 JSON 数据可以直接 parse，而避免用 cheerio 去解析主体内容的 HTML
+
+            const feedTitle = page.props.pageProps.articleContent.title;
+            const feedDesc = page.props.pageProps.articleContent.description;
+            const $author = page.props.pageProps.articleContent.author;
+            const authorName = $author.name;
+
+            const $blocks = page.props.pageProps.articleContent.blocks;
+
+            const anchorDay = dayjs();
+            let heading = null,
+                articleObj = {};
+            let year = anchorDay.year();
+            let prevMonth = -1;
+
+            $blocks.forEach((block) => {
+                const text = (block.text || '').trim();
+                if (!text) {
+                    return;
+                }
+                if (block.type === 'subheading') {
+                    if (heading !== null) {
+                        articleObj.description = articleObj.description.trim().replace(/\n/g, '<br/>');
+                        returns.push(articleObj);
+                        articleObj = {};
+                    }
+
+                    heading = text;
+
+                    articleObj.title = heading;
+                    articleObj.author = authorName;
+                    articleObj.category = 'ChatGPT';
+                    articleObj.link = articleUrl + '#' + block.idAttribute;
+                    articleObj.guid = articleUrl + '#' + block.idAttribute;
+                    articleObj.description = '';
+
+                    // 目前 ChatGPT Release Notes 页面并没有写入年份，所以只能靠猜
+                    // 当前的正则表达式只支持 (月份英文+空格+日期数字) 的格式
+                    const matchesPubDate = heading.match(/\((\w+\s+\d{1,2})\)$/);
+                    // 实现：当年度交替时，年份减去 1
+                    if (matchesPubDate !== null) {
+                        const curMonth = 1 + 'Jan,Feb,Mar,Apr,May,Jun,Jul,Aug,Sep,Oct,Nov,Dec'.split(',').indexOf(matchesPubDate[1].substring(0, 3));
+                        if (prevMonth !== -1 && prevMonth < curMonth) {
+                            year--; // 年度交替：上一个月份数小于当前月份数；但排除 prevMonth==-1 的初始化情况
+                        }
+
+                        prevMonth = curMonth;
+                        const pubDay = dayjs(`${matchesPubDate[1]}, ${year}`, ['MMMM D, YYYY', 'MMM D, YYYY'], 'en', true);
+                        // 从 ISO（GMT）时间的字符串（使用字符串替换的方式）替换成 US/Pacific PST 时区的表达
+                        articleObj.pubDate = dayjs(pubDay.toISOString().replace(/\.\d{3}Z$/, '-08:00'));
+
+                        const linkAnchor = pubDay.format('YYYY_MM_DD');
+                        articleObj.guid = articleUrl + '#' + linkAnchor;
+                    }
+                } else {
+                    articleObj.description += block.text.trim() + '\n\n';
+                }
+            });
+
+            if (heading !== null) {
+                articleObj.description = articleObj.description.trim().replace(/\n/g, '<br/>');
+                returns.push(articleObj);
+            }
+
+            return { feedTitle, feedDesc, items: returns };
+        },
+        config.cache.routeExpire,
+        false
+    );
+
+    ctx.state.data = {
+        title: cache.feedTitle,
+        description: cache.feedDesc,
+        link: articleUrl,
+        item: cache.items,
+    };
+};

package/lib/v2/openai/common.js

@@ -0,0 +1,72 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const { art } = require('@/utils/render');
+const path = require('path');
+
+const getApiUrl = async () => {
+    const blogRootUrl = 'https://openai.com/blog';
+
+    // Find API base URL
+    const initResponse = await got({
+        method: 'get',
+        url: blogRootUrl,
+    });
+
+    const apiBaseUrl = initResponse.data
+        .toString()
+        .match(/(?<=TWILL_API_BASE:").+?(?=")/)[0]
+        .replaceAll('\\u002F', '/');
+
+    return new URL(apiBaseUrl);
+};
+
+const parseArticle = (ctx, rootUrl, attributes) =>
+    ctx.cache.tryGet(attributes.slug, async () => {
+        const textUrl = `${rootUrl}/${attributes.slug}`;
+        const detailResponse = await got({
+            method: 'get',
+            url: textUrl,
+        });
+        let content = cheerio.load(detailResponse.data);
+
+        const authors = content('[aria-labelledby="metaAuthorsHeading"] > li > a > span > span')
+            .toArray()
+            .map((entry) => content(entry).text())
+            .join(', ');
+
+        // Leave out comments
+        const comments = content('*')
+            .contents()
+            .filter(function () {
+                return this.nodeType === 8;
+            });
+        comments.remove();
+
+        content = content('#content');
+
+        const imageSrc = attributes.seo.ogImageSrc;
+        const imageAlt = attributes.seo.ogImageAlt;
+
+        const article = art(path.join(__dirname, 'templates/article.art'), {
+            content,
+            imageSrc,
+            imageAlt,
+        });
+
+        // Not all article has tags
+        attributes.tags = attributes.tags || [];
+
+        return {
+            title: attributes.title,
+            author: authors,
+            description: article,
+            pubDate: attributes.createdAt,
+            category: attributes.tags.map((tag) => tag.title),
+            link: textUrl,
+        };
+    });
+
+module.exports = {
+    getApiUrl,
+    parseArticle,
+};

package/lib/v2/openai/maintainer.js

@@ -0,0 +1,4 @@
+module.exports = {
+    '/blog/:tag?': ['StevenRCE0', 'nczitzk'],
+    '/research': ['yuguorui'],
+};